Bio: Anna Schneider is a Data Science Manager on the Merch Algorithms team at Stitch Fix. Her team uses ML, OR, HCI, experimentation, and more to recommend how Stitch Fix sources half its inventory. Past lives include co-founding a clean energy data startup, earning a PhD in biophysics from UC Berkeley, and co-organizing the Unconscious Bias Project.
Talk: Just ask: Designing intent-driven algos
Abstract: Classic recommender systems are great for answering the question “what does a user want in general?”. However, they only get you partway to an answer to “what does a user want right now?”. To close the gap, it helps to capture and act on explicit user intent. I’ll share examples of this paradigm at Stitch Fix, and the resulting changes to our algorithms and architectures.